Research Article ISSN 2576-0564

Research ArticleInternational Journal of Microbiology & Infectious Diseases

Production of External Reference Materials in Food Microbiology

Marie de BORT*, Elvire MESSINEO, Romain LE NEVE, Abdelkader BOUBETRA and Anne TIRARD

Correspondence:
*

Marie de Bort, Bipea 189 rue d’Aubervilliers 75018 Paris, France,

Bipea 189 rue d’Aubervilliers 75018 Paris, France. E-mail: mdebort@bipea.org.

Received: 18 December 2017; Accepted: 24 January 2018

Citation: Marie de BORT, Elvire MESSINEO, Romain LE NEVE, et al. Production of External Reference Materials in Food
Microbiology. Int J Microbiol Infect Dis. 2018; 2(1): 1-3.

ABSTRACT
In order to help laboratories to face with their regulatory requirements, BIPEA (Bureau Interprofessionnel
d’Études Analytiques) developed a production of external reference materials (ERM) for microbiology in food.
These samples should allow laboratories to check the trueness of their results at any time, outside the regular
proficiency-testing schemes (PTS).

For this purpose, stable and homogeneous samples of minced meat spiked with various bacterial strains are
produced by BIPEA: Escherichia coli, Clostridium perfringens, Bacillus cereus, Staphylococcus aureus and
Listeria monocytogenes, at levels between 102 to 104 microorganisms per gram. Controls were performed by
analysing samples all along the production process, at the beginning of the study and regularly during a 6-month
period.

The homogeneity was checked by calculating coefficients of variation, which were inferior to 25 % for all the
analytical parameters. The stability was characterized by comparing means of three samples to the mean obtained
at the beginning of the study. The samples produced were thus considered as being sufficiently homogeneous and
stable to meet the ERM requirements: the results of enumeration of the different micro-organisms present in the
minced meat after 6 months of storage at (-24 ± 6)°C showed a good stability, with a maximum deviation less than
0.5 log (CFU/g: colony forming unity per gram).

Keywords for the development of reference materials for food microbiology

Stability, Homogeneity, External reference materials, Proficiency-
testing schemes, Microbiology.
Introduction
The development of microbiological reference materials has a
long history. In 1965 Kampelmacher was already using artificially
inoculated minced meat samples in comparative studies [1] as
a standard sample for the evaluation of the performance of the
Salmonella detection procedure in various laboratories. Later
Salmonella was dried onto milk powder which resulted in the
production of a reference material (gelatin capsules containing
spray dried milk artificially contaminated with Salmonella) for
use in laboratories methods trials [2]. The development of other
reference materials and their evaluation was initiated in 1986 when
the first contract for water microbiology with the former European
Communities Bureau of Reference (BCR) was agreed. A contract
Int J Microbiol Infect Dis, 2018
followed in 1987. These reference materials are prepared from
spray dried artificially contaminated milk. The initial spray dried
powder (called highly contaminated milk powder or HCMP)
is mixed with sterile milk powder to give the desired level of
contamination and the mixed powder is subsequently packed into
gelatin capsules.
Many regular proficiency-testing schemes offer real sample in
microbiology but do not allow laboratories to verify the trueness
of their results at any time. This is why BIPEA has developed new
ERMs, particularly in the field of food microbiology.
The external reference materials, of BIPEA are a production
surplus, consider as stable and homogenous enough and for which
an assigned value has been estimated. The ERMs could have
several uses: i) qualification of operators for laboratories accredited
Volume 1 | Issue 2 | 1 of 3
according to the requirements of ISO standard IEC 17025 [3], ii)
validation of alternative methods according to ISO 16140 standard
[4] or internal methods, iii) determination of reproducibility and
repeatability of reference methods [5-7]. This article describes the
different stages of the implementation of ERMs in minced meat
matrix spiked with several bacterial strains, after storage at (-24
± 6)°C.
Experimental design
Sterile minced meat, divided into 25 grams samples, was spiked
with several bacterial strains at given concentrations. Three
of these samples were analyzed each month over a period of 6
months. To be used as ERMs, it is required to demonstrate the
homogeneity between these samples and their stability over time,
after storage at (-24 ± 6)°C.
Materials and Methods
Materials
Five bacterial strains were used for the contamination of minced
meat matrix, Escherichia coli, Clostridium perfringens, Bacillus
cereus, Staphylococcus aureus and Listeria monocytogenes. In
the case of Clostridium perfringens and Bacillus cereus, these
strains were used in their sporulated form in order to increase their
stability.
Methods
The minced meat was divided into 25 grams samples in sterile
stomacher bags. These bags were then sealed and sterilized by
ionization. The aim of this ionization is to sanitize microbiologically
the matrix in order to avoid interactions between the target
microorganisms and the endogenous flora.
Each bag was then inoculated individually from a pool of bacterial
strains suspended in specific diluent. The target concentrations of
the spiking suspensions were obtained by carrying out dilutions
from the primary suspensions.
The samples were spiked at levels close to 103 CFU/g for all
bacterial species tested except for Clostridium perfringens which
was inoculated at a rate of about 2.102 CFU/g and Bacillus cereus
at a rate of 5.102 CFU/g. The concentrations used reflect the rates
generally observed in this type of matrix.
The samples were analyzed at different time intervals (at D0
and then each month during six months) according to the
reference methods for each analytical parameter (Total viable
count, Escherichia coli, Coliforms, Enterobacteria, Clostridium
perfringens, Bacillus cereus, Staphylococcus aureus and Listeria
monocytogenes).
Results and Discussions
Rate of recovery
At D0, which corresponds to the day of preparation of the spiked
samples, the data show that the recovery rate reaches 100% for
all strains species tested except for Listeria monocytogenes
strains, which recovered at a 90% rate. These results show that
Int J Microbiol Infect Dis, 2018
the recovery rate for all microorganisms is satisfactory to start the
trials (Table 1).
Coefficient of variation – Homogeneity
The coefficients of variation (CVs) [8] in the study show the
reliability that can be attributed to the assigned values which are
compared to assess the stability of the samples. From an analytical
point of view, the coefficients of variation allow to sorting out
which analysis is under control and which raises difficulties.
The coefficients of variation observed (table 1) range from 0.5%
to 22.2%. The inter-samples variation is therefore sufficiently
low (CVs less than 25%) to conclude that these samples are
homogeneous regarding to the variability of the analytical methods
and the matrix composition.
Stability
One of the aims of this study is, in addition to the development
of homogeneous samples, to determine over which period of time
these samples can remain stable. Since microorganisms are living
organisms, their concentration can be expected to evolve over time.
It may increase, if the elements necessary for their multiplication
are present in the matrix used with favorable conditions in terms
of temperature and aero philia, or it may on the contrary decrease
due to the death of these microorganisms [9]. However, some
techniques such as freezing can preserve their concentration [10].
Indeed, freezing has the effect of slowing down, or even almost
halting, the processes allowing them to live and develop without
causing their death. Therefore, it was decided to store the samples
at (-24 ± 6)°C, in order to stabilize as much as possible, the
concentration of the bacterial strains within the minced meat.
The results of the different analyses show that the difference of
enumeration of microorganisms between the analysis of these
samples at D0 and their analysis after several months does not
vary by more than 0.5 log. This data confirms that the enumeration
of microorganisms in minced meat is stable after six months of
freezing.
Conclusions
The results of the counting of the various microorganisms present
in the minced meat after 6 months of storage at (-24 ± 6)°C show
satisfactory stability of the samples with a difference of extreme
values less than 0.5 log for all the analytical parameters. The
homogeneity of the samples is also satisfactory with coefficients
of variation below 25% for all the analytical parameters.
The results of homogeneity and stability of the samples are therefore
in accordance with those expected: satisfactory homogeneity and
stability for 6 months at (-24 ± 6)°C.
This study enabled the implementation of a test production for
trustworthy ERM’s storage and shipping conditions.
Eight analytical parameters can be studied at any time by the
laboratories to check their performance: total viable count,
Volume 1 | Issue 2 | 2 of 3
 D0 M1 M2 M3 M4 M5 M6
mean concentration [CFU/g] 3.6E+03 3.2E+03 2.7E+03 2.5E+03 1.8E+03 1.7E+03 1.6E+03
Total viable Count mean concentration (log[CFU/g]) 3.56 3.51 3.43 3.40 3.26 3.23 3.20
Coefficient of variation (%) 3.9 4.4 15.3 11.6 14.8 0.5 2.2
mean concentration [CFU/g] 1.1E+03 9.1E+02 7.5E+02 6.7E+02 5.7E+02 4.4E+02 4.4E+02
Escherichia coli mean concentration (log[CFU/g]) 3.04 2.96 2.88 2.83 2.76 2.64 2.64
Coefficient of variation (%) 7.5 5.7 5.6 7.5 9.7 5.2 12.5
mean concentration [CFU/g] 1.0E+03 8.8E+02 8.2E+02 6.9E+02 5.1E+02 4.3E+02 3.5E+02
Coliforms mean concentration (log[CFU/g]) 3.00 2.94 2.91 2.84 2.71 2.63 2.54
Coefficient of variation (%) 8.4 6.4 7.1 3.8 14.3 5.6 11.8
mean concentration [CFU/g] 1.0E+03 8.5E+02 8.1E+02 7.5E+02 4.8E+02 4.5E+02 3.6E+02
Enterobacteria mean concentration (log[CFU/g]) 3.00 2.93 2.91 2.88 2.68 2.65 2.56
Coefficient of variation (%) 5.6 3.2 8.5 9.7 14.7 5.4 7.8
mean concentration [CFU/g] 2.0E+02 2.0E+02 2.2E+02 2.0E+02 2.2E+02 2.0E+02 1.8E+02
Clostridium perfringens mean concentration (log[CFU/g]) 2.30 2.30 2.34 2.30 2.34 2.30 2.26
Coefficient of variation (%) 5.3 9.1 16.6 10.3 8.3 11.3 7.9
mean concentration [CFU/g] 5.1E+02 4.9E+02 4.0E+02 3.6E+02 3.1E+02 2.7E+02 2.4E+02
Bacillus cereus mean concentration (log[CFU/g]) 2.71 2.69 2.60 2.56 2.49 2.43 2.38
Coefficient of variation (%) 4.1 9.4 2.3 3.9 1.6 3.0 7.7
mean concentration [CFU/g] 1.1E+03 9.4E+02 7.4E+02 6.6E+02 5.2E+02 4.9E+02 5.0E+02
Staphylococcus aureus mean concentration (log[CFU/g]) 3.04 2.97 2.87 2.82 2.72 2.69 2.70
Coefficient of variation (%) 4.8 7.7 7.8 3.5 21.5 6.7 9.3
mean concentration [CFU/g] 9.3E+02 8.7E+02 6.9E+02 6.9E+02 5.0E+02 5.0E+02 5.0E+02
Listeria monocytogenes mean concentration (log[CFU/g]) 2.97 2.94 2.84 2.84 2.70 2.70 2.70
2.9 5.8 6.0 8.9 22.2 15.0 8.5
Table 1: Results obtained from the analysis of samples of minced meat contaminated with several bacterial strains over a period of 6 months after
storage at (-24 ± 6)°C.
Escherichia coli, Coliforms, Enterobacteria, Clostridium reference method.
perfringens, Bacillus cereus, Staphylococcus aureus and Listeria 5. International standard: ISO 5725-1:1994 - Accuracy trueness
monocytogenes. The use of ERM is one of the important tool for and precision of measurement methods and results -- Part 1:
the monitoring of quality assurance and should for example allow General principles and definitions.
methods validation and operators’ qualification using a real food 6. International standard: ISO 5725-2:1994 - Accuracy trueness
matrix. and precision of measurement methods and results -- Part
2: Basic method for the determination of repeatability and
References reproducibility of a standard measurement method.
1. Kampelmacher EH. Comparative studies into the isolation of 7. International standard: ISO 5725-6:1994 - Accuracy trueness
Salmonella from minced meat in five laboratories. Zbl. Bakt. and precision of measurement methods and results -- Part 6:
I. Abt. Orig. 1967; 204: 100-111. Use in practice of accuracy values .
2. Beckers HJ, van Leusden FM, Meijssen MJM, et al. Reference 8. Reed GF, Lynn F, Meade BD. Use of coefficient of variation in
material for the evaluation of the standard method for the assessing variability of quantitative assays. Clin. Diagn. Lab.
detection of salmonellas in foods and feeding stuffs. J. Appl. Immunol. 2002; 9: 1235-1239.
Bacteriol. 1985; 59: 507-512. 9. Mihoub F, Mistou MY, Guillot A, et al. Cold adaptation of
3. International standard: ISO/IEC 17025:2005 - General Escherichia coli: microbiological and proteomic approaches.
requirements for the competence of testing and calibration Int. J. Food. Microbiol. 2003; 89: 171-184.
laboratories. 10. El-Kest SE, Marth EH. Injury and death of frozen Listeria
4. International standard: ISO 16140-2:2016 - Microbiology of monocytogenes as affected by glycerol and milk components.
the food chain -- Method validation -- Part 2: Protocol for J. Dairy Sci. 1991; 74: 1201-1208.
the validation of alternative proprietary methods against a

© 2018 Marie de BORT, et al. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License

Int J Microbiol Infect Dis, 2018 Volume 1 | Issue 2 | 3 of 3